Paper Citation Record · LEDGER

How do LLMs Compute Verbal Confidence

As of 6 August 2026, this Paper Citation Record lists 29 of 29 outbound references and 11 inbound Pith citation observations for arXiv:2603.17839.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2603.17839 v3

Coverage vector

measured 29 of 29 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-05-21T10:38:28.533485Z

measured 40 of 40 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-06T06:34:29.942622+00:00

measured 11 of 11 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-02T16:59:54.843386Z

measured 1 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: pith, observed 2026-08-05T02:28:24.338817Z

Reference resolution

29 of 29 outbound references displayed

verified exact17
verified fuzzy6
unresolved1
parse uncertain0
malformed identifier1
metadata mismatch4

External citation measurements

0
pith, observed 2026-08-05T02:28:24.338817Z

Outbound references

Observation 2071036b-effb-4c3b-9515-523b916c5904 · outbound

This paper cites Anthropic.

How do LLMs Compute Verbal Confidence Anthropic

Reference 1

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T10:40:01.036734Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation d291969d-3eca-4093-a1d0-3903786e1bcf · outbound

This paper cites The Internal State of an LLM Knows When It's Lying.

How do LLMs Compute Verbal Confidence The Internal State of an LLM Knows When It's Lying

Reference 2

Resolution: verified exact; local_arxiv, observed 2026-05-21T10:40:00.625982Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation d64fab97-26cc-4a83-84bc-4d05345c50ef · outbound

This paper cites Discovering Latent Knowledge in Language Models Without Supervision.

How do LLMs Compute Verbal Confidence Discovering Latent Knowledge in Language Models Without Supervision

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-05-21T10:40:00.611725Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation ca5e39d1-085f-4f68-b494-aef04b538cff · outbound

This paper cites Trace length is a simple un- certainty signal in reasoning models.arXiv preprint arXiv:2510.10409.

How do LLMs Compute Verbal Confidence Trace length is a simple un- certainty signal in reasoning models.arXiv preprint arXiv:2510.10409

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.629780Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 978a1163-3b6b-4158-87b3-e9c28d4fdfa3 · outbound

This paper cites A Survey of Confidence Estimation and Calibration in Large Language Models.

How do LLMs Compute Verbal Confidence A Survey of Confidence Estimation and Calibration in Large Language Models

Reference 5

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.622154Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation ee1770bc-8d1b-4947-b9cb-4488a8f09bd7 · outbound

This paper cites Dissecting Recall of Factual Associations in Auto-Regressive Language Models.

How do LLMs Compute Verbal Confidence Dissecting Recall of Factual Associations in Auto-Regressive Language Models

Reference 6

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.615877Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 43822381-f42d-4b53-9996-33fc130ef8ed · outbound

This paper cites How to use and interpret activation patching.

How do LLMs Compute Verbal Confidence How to use and interpret activation patching

Reference 7

Resolution: verified exact; local_arxiv, observed 2026-05-21T10:40:00.618738Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation f9a67b9d-08e2-4efb-b1d7-2a7999e578d5 · outbound

This paper cites arXiv preprint arXiv:2510.20487 , year =.

How do LLMs Compute Verbal Confidence arXiv preprint arXiv:2510.20487 , year =

Reference 8

Resolution: metadata mismatch; arxiv_id, observed 2026-05-21T10:40:00.633194Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 5b0350cc-9965-4152-b816-eeaaf29d7aae · outbound

This paper cites TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension.

How do LLMs Compute Verbal Confidence TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehension

Reference 9

Resolution: metadata mismatch; local_arxiv, observed 2026-05-21T10:40:00.545582Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 713ddb4c-558b-47cc-8bbd-9220b569bbf0 · outbound

This paper cites Language Models (Mostly) Know What They Know.

How do LLMs Compute Verbal Confidence Language Models (Mostly) Know What They Know

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-05-21T10:40:00.541936Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation e3685beb-482a-4ea2-82da-32eda5d7a4d8 · outbound

This paper cites Liu, J., Jain, J., Diab, M., and Subramani, N.

How do LLMs Compute Verbal Confidence Liu, J., Jain, J., Diab, M., and Subramani, N

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.589501Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 2a7e6c72-cb4d-4321-8423-576de713d3e0 · outbound

This paper cites Reasoning about Uncertainty: Do Reasoning Models Know When They Don't Know?.

How do LLMs Compute Verbal Confidence Reasoning about Uncertainty: Do Reasoning Models Know When They Don't Know?

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.608508Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 91505d07-510c-45a2-b16d-47e4c7da2b3f · outbound

This paper cites Steering Llama 2 via Contrastive Activation Addition.

How do LLMs Compute Verbal Confidence Steering Llama 2 via Contrastive Activation Addition

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-05-21T10:40:00.549257Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 750c528e-5565-4b2c-82cd-78c5421b8f5e · outbound

This paper cites A practical review of mechanistic interpretability for transformer-based language models.arXiv preprint arXiv:2407.02646.

How do LLMs Compute Verbal Confidence A practical review of mechanistic interpretability for transformer-based language models.arXiv preprint arXiv:2407.02646

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.600898Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation bd53ed0b-2a70-46e9-be04-93410ef94825 · outbound

This paper cites Generalization of Fine-Tuned Uncertainty Communication and Metacognition in Large Language Models.

How do LLMs Compute Verbal Confidence Generalization of Fine-Tuned Uncertainty Communication and Metacognition in Large Language Models

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-06-23T02:11:58.848996Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation e571b856-0a57-4b15-bfc3-fc1a4353dd58 · outbound

This paper cites Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback.

How do LLMs Compute Verbal Confidence Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback

Reference 16

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.577255Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 67eb0537-02dc-4c9c-93d3-cb0fc4e758fe · outbound

This paper cites Steering Language Models With Activation Engineering.

How do LLMs Compute Verbal Confidence Steering Language Models With Activation Engineering

Reference 17

Resolution: metadata mismatch; local_arxiv, observed 2026-05-21T10:40:00.604683Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 8f6a15ae-07e4-4c1a-acb0-6a4def1411cd · outbound

This paper cites A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation.

How do LLMs Compute Verbal Confidence A Stitch in Time Saves Nine: Detecting and Mitigating Hallucinations of LLMs by Validating Low-Confidence Generation

Reference 18

Resolution: metadata mismatch; arxiv_id, observed 2026-05-21T10:40:00.596553Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 7f4f41c3-ab0d-4596-85b5-9238bc1bff32 · outbound

This paper cites Base Models Know How to Reason, Thinking Models Learn When.

How do LLMs Compute Verbal Confidence Base Models Know How to Reason, Thinking Models Learn When

Reference 19

Resolution: verified exact; arxiv_id, observed 2026-07-08T02:18:39.926237Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation f2317251-b46c-4673-939f-1e3c263a893d · outbound

This paper cites Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs.

How do LLMs Compute Verbal Confidence Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-05-21T10:40:00.556530Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 576d8437-c0e4-41f9-a7b7-e7f87ac062ff · outbound

This paper cites Reasoning models better express their confidence.arXiv preprint arXiv:2505.14489.

How do LLMs Compute Verbal Confidence Reasoning models better express their confidence.arXiv preprint arXiv:2505.14489

Reference 21

Resolution: verified exact; arxiv_id, observed 2026-05-21T10:40:00.553191Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 614b9f79-9a84-4afe-911f-0318a4031e04 · outbound

This paper cites Towards Best Practices of Activation Patching in Language Models: Metrics and Methods.

How do LLMs Compute Verbal Confidence Towards Best Practices of Activation Patching in Language Models: Metrics and Methods

Reference 22

Resolution: verified exact; local_arxiv, observed 2026-05-21T10:40:00.592797Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 804c98de-4109-43b9-9d3a-3a2768db66a0 · outbound

This paper cites Almost certain.

How do LLMs Compute Verbal Confidence Almost certain

Reference 23

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T10:40:01.034622Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 05d8ca97-384e-4cc2-ae1b-daaa9e29639a · outbound

This paper cites The model’s performance was 77.4%; this was determined by having GPT4o-mini mark questions (B) Distribution of Gemma’s confidence responses across the 10 classes.

How do LLMs Compute Verbal Confidence The model’s performance was 77.4%; this was determined by having GPT4o-mini mark questions (B) Distribution of Gemma’s confidence responses across the 10 classes

Reference 24

Resolution: malformed identifier; raw_fallback, observed 2026-05-21T10:40:01.032617Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation cb8dd214-f650-4fe3-a512-b64dd825c513 · outbound

This paper cites 20 How do LLMs Compute Verbal Confidence? Figure 15.Calibration and Distribution of Categorical Confidence Ratings in Qwen 2.5 7b.

How do LLMs Compute Verbal Confidence 20 How do LLMs Compute Verbal Confidence? Figure 15.Calibration and Distribution of Categorical Confidence Ratings in Qwen 2.5 7b

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T10:40:01.030450Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 90194fb4-a870-4988-8bf2-392f4eb57367 · outbound

This paper cites Almost certain.

How do LLMs Compute Verbal Confidence Almost certain

Reference 26

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T10:40:01.028082Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation be8370be-8779-434d-b185-262064d09d91 · outbound

This paper cites an unresolved cited work.

How do LLMs Compute Verbal Confidence Unresolved cited work

Reference 27

Resolution: unresolved; raw_fallback, observed 2026-05-21T10:40:01.025444Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 1632c308-2542-4453-9dc3-54f027f3d327 · outbound

This paper cites Highly likely.

How do LLMs Compute Verbal Confidence Highly likely

Reference 28

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T10:40:01.020838Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 7547dc66-51ad-4a03-aa15-dea2275643ee · outbound

This paper cites Logprobs explained only 4.9% of variance in within-run verbal con- fidence (r= 0.23 , R2 CV = 0.049 ) and 8.4% in cross-run verbal confidence (r= 0.29 , R2 CV = 0.084).

How do LLMs Compute Verbal Confidence Logprobs explained only 4.9% of variance in within-run verbal con- fidence (r= 0.23 , R2 CV = 0.049 ) and 8.4% in cross-run verbal confidence (r= 0.29 , R2 CV = 0.084)

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-05-21T10:40:01.023150Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Pith citing papers

Observation 917d2708-72ca-4af2-8611-9e7cfb4de52e · inbound

Wired for Overconfidence: A Mechanistic Perspective on Inflated Verbalized Confidence in LLMs cites this paper.

Wired for Overconfidence: A Mechanistic Perspective on Inflated Verbalized Confidence in LLMs How do LLMs Compute Verbal Confidence

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-02T16:59:54.843386Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c9e63885-a2d9-4606-9dd9-ee75b5140df0 · inbound

Verbal Confidence Saturation in 3-9B Open-Weight Instruction-Tuned LLMs: A Pre-Registered Psychometric Validity Screen cites this paper.

Verbal Confidence Saturation in 3-9B Open-Weight Instruction-Tuned LLMs: A Pre-Registered Psychometric Validity Screen How do LLMs Compute Verbal Confidence

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-05-20T02:04:53.698588Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 1ba2f18b-f0fa-4d29-9862-6f8c88201457 · inbound

How LLMs Detect and Correct Their Own Errors: The Role of Internal Confidence Signals cites this paper.

How LLMs Detect and Correct Their Own Errors: The Role of Internal Confidence Signals How do LLMs Compute Verbal Confidence

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-05-20T02:04:53.698588Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation bda01001-2017-4f8c-8a4a-b956aa20f792 · inbound

Distilling Self-Consistency into Verbal Confidence: A Pre-Registered Negative Result and Post-Hoc Rescue on Gemma 3 4B cites this paper.

Distilling Self-Consistency into Verbal Confidence: A Pre-Registered Negative Result and Post-Hoc Rescue on Gemma 3 4B How do LLMs Compute Verbal Confidence

Reference 9

Resolution: verified exact; arxiv_id, observed 2026-05-20T02:04:53.698588Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation cab58a57-38f9-4728-8732-bf73168dec87 · inbound

Hypothesis generation and updating in large language models cites this paper.

Hypothesis generation and updating in large language models How do LLMs Compute Verbal Confidence

Reference 25

Resolution: verified exact; arxiv_id, observed 2026-05-20T02:04:53.698588Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation bb55f543-1bdb-4cb5-8802-838fd3379d05 · inbound

Internalizing Curriculum Judgment for LLM Reinforcement Fine-Tuning cites this paper.

Internalizing Curriculum Judgment for LLM Reinforcement Fine-Tuning How do LLMs Compute Verbal Confidence

Reference 35

Resolution: verified exact; arxiv_id, observed 2026-05-20T02:04:53.698588Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 08872acb-0634-4d40-800e-2e2f34f65117 · inbound

The Hidden Signal of Verifier Strictness: Controlling and Improving Step-Wise Verification via Selective Latent Steering cites this paper.

The Hidden Signal of Verifier Strictness: Controlling and Improving Step-Wise Verification via Selective Latent Steering How do LLMs Compute Verbal Confidence

Reference 6

Resolution: verified exact; local_arxiv, observed 2026-05-21T06:54:01.045732Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation ded4290c-c916-45a7-8b32-8946cfaa240c · inbound

CALIBER: Calibrating Confidence Before and After Reasoning in Language Models cites this paper.

CALIBER: Calibrating Confidence Before and After Reasoning in Language Models How do LLMs Compute Verbal Confidence

Reference 10

Resolution: verified exact; local_arxiv, observed 2026-07-04T16:39:58.391902Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 2be82f7a-b8e1-4ce7-877e-7a3c5de4f199 · inbound

Reported Confidence in LLMs Tracks Commitment More Than Correctness cites this paper.

Reported Confidence in LLMs Tracks Commitment More Than Correctness How do LLMs Compute Verbal Confidence

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-06-30T08:04:28.800763Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-06T06:34:29.942622+00:00.

Observation 98f1e473-ae9b-4673-89ca-7236a8042990 · inbound

The Computational Basis of Confidence in Large Language Models cites this paper.

The Computational Basis of Confidence in Large Language Models How do LLMs Compute Verbal Confidence

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-02T06:37:39.945125Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5440aae2-db78-4d44-86ed-c04233837214 · inbound

Thinking Under Uncertainty: Evidence Use and Information-Seeking in Language Models cites this paper.

Thinking Under Uncertainty: Evidence Use and Information-Seeking in Language Models How do LLMs Compute Verbal Confidence

Reference 7

Resolution: unresolved; no resolver link, observed 2026-07-30T19:24:33.900945Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.